Let Features Decide for Themselves: Feature Mask Network for Person Re-identification
This addresses the problem of accurate identity matching in multi-camera systems for security and surveillance applications, representing an incremental advance with specific performance gains.
The paper tackles person re-identification by proposing a Feature Mask Network that uses high-level features to predict masks for reweighting low-level features, achieving improvements of 5.3%, 9.1%, and 10.7% in mAP on three datasets compared to state-of-the-art methods.
Person re-identification aims at establishing the identity of a pedestrian from a gallery that contains images of multiple people obtained from a multi-camera system. Many challenges such as occlusions, drastic lighting and pose variations across the camera views, indiscriminate visual appearances, cluttered backgrounds, imperfect detections, motion blur, and noise make this task highly challenging. While most approaches focus on learning features and metrics to derive better representations, we hypothesize that both local and global contextual cues are crucial for an accurate identity matching. To this end, we propose a Feature Mask Network (FMN) that takes advantage of ResNet high-level features to predict a feature map mask and then imposes it on the low-level features to dynamically reweight different object parts for a locally aware feature representation. This serves as an effective attention mechanism by allowing the network to focus on local details selectively. Given the resemblance of person re-identification with classification and retrieval tasks, we frame the network training as a multi-task objective optimization, which further improves the learned feature descriptions. We conduct experiments on Market-1501, DukeMTMC-reID and CUHK03 datasets, where the proposed approach respectively achieves significant improvements of $5.3\%$, $9.1\%$ and $10.7\%$ in mAP measure relative to the state-of-the-art.